Trees produce various types of wood timber for industry and society construction. One of the most valuable and popular types of wood, especially in China, is the natural durable wood with splendid colours. Natural durable wood has self-preservation ability to survive biological decay [1, 2]. Besides, different wood colours, for instance, yellow and red, potentially increase the wood value for end use. Therefore, natural durable wood with colours has been widely studied. In the tree stem, the inner part of wood is called heartwood, and the outside of wood is recognized as sapwood. As the tree grows, cells in the inner part of sapwood begin to die and accumulate massive secondary metabolites. Meanwhile, the sapwood turns into heartwood with natural durability and colour [3]. However, traditional methods testing wood natural durability are costly and time-consuming. It is reported that the extractives in the heartwood play an important role in the formation of colour and natural durability [4]. Decomposition and discolouration will occur when the extractives are removed from the durable wood [5]. Therefore, the amount of EC in heartwood has been studied as a proxy for natural durability [6].
The variation of EC in heartwood is huge and can be decreased by genetic selection [7]. There are different ways to determine the quantity of extractives in heartwood. Traditional methods such as Soxhlet and accelerated solvent extraction (ASE) [8] are time- and cost-consuming and not suitable for tree breeding and selection programs, which rely on the measurement of large number of samples. Therefore, a high throughput and rapid measurement method for EC is needed.
Near infrared spectroscopy (NIR) is a non-destructive technique that is used for the analysis of the composition of chemical compounds in general [9-11]. It is applied to determine the quantity of heartwood extractives in some tree species [12-14] and it yields promising and reliable results. The NIR spectra collected from different samples, either wood powder or solid, influence the performance of wood traits prediction model.
The models built on the different size of milled wood powder are different and perform higher accuracy than the model based on the solid wood when predicting chemical properties in eucalyptus wood with NIR spectroscopy [15]. However, wood sample grinding is also a time-consuming step and it reduces the extractive content prediction time. EC prediction from the solid wood samples is a suitable alterative way for NIR model calibration.
NIR spectra taken from the solid wood samples are influenced by many factors, such as moisture content [16] and grain angle [17]. It is claimed that the grain angle influences the EC prediction of Eucalyptus bosistoana with NIR spectra and this influence can be minimized by external parameter orthogonalization (EPO) algorithm [6]. Alternatively, this influence will be reduced by feature selection methods. One of the most important feature selection methods that can help NIR calibrations to get rid of confounding effects [18-20] is called significant Multivariate Correlation (sMC) [21]. It is aimed to find the most important variables in NIR spectra and can remove the irreverent variables that influence the accuracy of the model prediction when predicting the target chemicals content in plants. It is a good choice to conduct the method of feature selection combined with partial least squares regression (PLS). However, the important features in the NIR spectra for the grain angles and EC are little known.
T. sinensis is a native Chinese tree species that has been widely distributed in China. It has a long history of cultivation for its digestible buds in China. In addition, T. sinensis also has the advantages of fast growing and bright red heartwood which lead to its wide use in furniture and industry. It is widely studied in many fields, including the cultivation, reproduction, biological activity of T. sinensis, and physical and chemical characteristics of its wood. However, little is known about the nature durability and red colour timber. To establish a high quality breeding program for durability and red wood selection, an alternative way is needed to allow a fast and efficient measurement of the heartwood quality of T. sinensis. If NIR can be successfully used to analyse the heartwood properties without any grain angle influence, traditional methods which are time- and cost-consuming can be replaced. And consequently, it will give a lot of benefit for selection.
Hence, This study will focus on the effect of grain angle on NIR spectra obtained from T. sinensis cores, and study the possibility of applying NIR as a rapid and precise method to predict the extractives content from the solid core samples of T. sinensis without grain angle influence.